You are on page 1of 62

Module 7 : Lecture 1


(Incompressible Flow Part I)
Accurate measurement in a flowing medium is always desired in many applications.
The basic approach of the given measurement technique depends on the flowing
medium (liquid/gas), nature of the flow (laminar/turbulent) and steady/unsteadiness of
the medium. Accordingly, the fluid flow diagnostics are classified as measurement of
local properties (velocity, pressure, temperature, density, viscosity, turbulent intensity
etc.), integrated properties (mass and volume flow rate) and global properties (flow
visualization). Also, these properties can be measured directly using certain devices or
can be inferred from few basic measurements. For instance, if one wishes to measure
the flow rate, then a direct measurement of volume/mass flow can be done during a
fixed time interval. However, the secondary approach is to measure some other
quantity such as pressure difference and/or fluid velocity at a point in the flow and
then calculate the flow rate using suitable expressions. In addition, flow-visualization
techniques are sometimes employed to obtain an image of the overall flow field. The










pressure/temperature, fluid velocity and its flow rate.

Measurement of Viscosity
The device used for measurement of viscosity is known as viscometer and it uses the
basic laws of laminar flow. The principles of measurement of some commonly used
viscometers are discussed here;
Rotating Cylinder Viscometer: It consists of two co-axial cylinders suspended coaxially as shown in the Fig. 7.1.1. The narrow annular space between the cylinders is
filled with a liquid for which the viscosity needs to be measured. The outer cylinder
has the provision to rotate while the inner cylinder is a fixed one and has the provision
to measure the torque and angular rotation. When the outer cylinder rotates, the torque
is transmitted to the inner stationary member through the thin liquid film formed
between the cylinders. Let r1 and

be the radii of inner and outer cylinders, h be the

depth of immersion in the inner cylinder in the liquid and


t r2

Page 1 of 56

is the annular


Page 2 of 56

gap between the cylinders. Considering N as the speed of rotation of the cylinder in
rpm, one can write the expression of shear stress

from the definition of viscosity

, as given below;

2r2 N

60 t



This shear stress induces viscous drag in the liquid that can be calculated by
measuring the toque through the mechanism provided in the inner cylinder.
2r2 N
T shear stressarearadius
2rh r

60 t


2 r1 2 2r hN CN
Here, C is a constant quantity for a given viscometer.

Fig. 7.1.1: Schematic nomenclature of a rotating cylinder viscometer.


Falling Sphere Viscometer: It consists of a long container of constant area filled with
a liquid whose viscosity has to be measured. Since the viscosity depends strongly with
the temperature, so this container is kept in a constant temperature bath as shown in
Fig. 7.1.2.

Fig. 7.1.2: Schematic diagram of a falling sphere viscometer.

A perfectly smooth spherical ball is allowed to fall vertically through the liquid
by virtue of its own weight W . The ball will accelerate inside the liquid, until
net downward force is zero i.e. the submerged weight of the ball


is equal to the

resisting force


given by Stokes law. After this point, the ball will move at

steady velocity which is known as terminal velocity. The equation of motion may be
written as below;

wl and


D wl FR


are the specific weights of the liquid and the ball, respectively. If

the spherical ball has the diameter D that moves at constant fall velocity V in a fluid
having viscosity , then using Stokes law, one can write the expression for resisting

force FR .


Substituting Eq. (7.1.4) in Eq. (7.1.3) and solving for ,



where V


The constant fall velocity can be calculated by measuring the time t taken by the
ball to fall through a distance


It should be noted here that the falling

sphere viscometer is applicable for the Reynolds number below 0.1 so that wall will
not have any effect on the fall velocity.
Capillary Tube Viscometer: This type of viscometer is based on laminar flow through
a circular pipe. It has a circular tube attached horizontally to a vessel filled with a
liquid whose viscosity has to be measured. Suitable head


is provided to the

liquid so that it can flow freely through the capillary tube of certain length

into a

collection tank as shown in Fig. 7.1.3. The flow rate Q of the liquid having specific

can be measured through the volume flow rate in the tank. The Hagen-

Poiseuille equation for laminar flow can be applied to calculate the viscosity of
the liquid.

wl hf d

128 QL

Fig. 7.1.3: Schematic diagram of a capillary tube viscometer.


Saybolt and Redwood Viscometer: The main disadvantage of the capillary tube
viscometer is the errors that arise due to the variation in the head loss and other
parameters. However, the Hagen-Poiseuille formula can be still applied by designing
a efflux type viscometer that works on the principle of vertical gravity flow of a
viscous liquid through a capillary tube. The Saybolt viscometer has a vertical
cylindrical chamber filled with liquid whose viscosity is to be measured (Fig. 7.1.4-a).
It is surrounded by a constant temperature bath and a capillary tube (length 12mm and
diameter 1.75mm) is attached vertically at the bottom of the chamber. For
measurement of viscosity, the stopper at the bottom of the tube is removed and time
for 60ml of liquid to flow is noted which is named as Saybolt seconds. So, Eq. (7.1.6)
can be used for the flow rate

is calculated by recording the time (Saybolt

seconds) for collection of 60ml of liquid in the measuring flask. For calculation
purpose of kinematic viscosity , the simplified expression is obtained as below;

0.002 t ; where, in Stokes and t in seconds


A Redwood viscometer is another efflux type viscometer (Fig. 7.1.4-b) that

works on the same principle of Saybolt viscometer. Here, the stopper is replaced with
an orifice and Redwood seconds is defined for collection of 50ml of liquid to flow out
of orifice. Similar expressions can be written for Redwood viscometer. In general,
both the viscometers are used to compare the viscosities of different liquid. So, the
value of viscosity of the liquid may be obtained by comparison with value of time for
the liquid of known viscosity.

Fig. 7.1.4: Schematic diagram: (a) Saybolt viscometer; (b) Redwood viscometer.

Measurement of Pressure
The fluid pressure is usually measured with reference to standard atmosphere (i.e.
760mm of mercury/101.325 kPa). Any differential pressures are often expressed as
gauge/vacuum pressure. The pressure measuring devices mostly employed in fluid
systems are generally grouped under two categories; liquid manometers and
mechanical gauges.
The liquid manometers work on the principle of balancing the column of liquid
whose pressure is to be determined by same/another liquid column. Depending on the
application, magnitude of pressure and sensitivity requirement, the manometers can
be selected. The most commonly used liquid as manometric fluid are mercury, water,
alcohol and kerosenes etc. Most of the case, for gauge pressure measurements,
mercury is widely used as manometric fluid because it has non-evaporating quality
under normal conditions, sharp meniscus and stable density. For some pressure
differences and low level vacuum, water can be considered as working fluid in the
manometer. Manometers can be employed to measure pressures in the range of 0.4 Pa
to 200 kPa.
The liquid manometers become bulky for handling higher pressure
measurements. In such cases, mechanical gauges are normally employed. These
gauges employ elastic elements which can deflect due to pressure acting on it. The
deflection obtained by action of pressure is mechanically magnified and made to
operate a pointer moving in a graduated dial. Some of these mechanical devices are
dead-weight pressure gauges, bourdon tube pressure gauge, elastic diaphragm
pressure gauges, pirani and McLeod gauges (vacuum measurement) etc.
Measurement of Temperature
Temperature is a thermodynamic property of a fluid which is measured as a change
with respect to another temperature-dependent property. In practical aspects, the
temperature is gauged by its effect on quantities such as volume, pressure, electrical
resistance and radiant energy. The temperature sensing devices working on these
techniques are classified in the following categories;

Thermometers (changes in physical dimension and gas/vapour pressures)

Resistance temperature

detectors (RTD), Thermistors, Thermocouples,

Semiconductor sensors (changes in electrical properties)

Pyrometers (Changes in thermal radiation)

Among all the devices, the electrical temperature sensors are mostly used
particularly when automatic/remote recording is desired. Radiant sensors are used for
noncontact temperature sensing, either in high temperature applications (combustors)
or for infrared sensing at low temperatures. These are optical devices and can be
adapted to whole-field temperature measurements known as thermal imaging. The
most familiar type of temperature sensing device is the thermometer that appears in
laboratories and households because of its ease in use and low cost. Some of the
important and commonly used temperature devices are discussed here.
Resistance Thermometers and Thermistors: Traditionally, the resistance elements
sensitive to temperatures are made out of metals which are good conductors of
electricity (e.g. nickel, platinum, copper and silver). The operating ranges for this
class of devices fall between 250C to 1000C. They are commonly referred as
resistance temperature detectors (RTD) and provide a linear temperature-resistance
R1c T T0


Here, R is the resistance at temperature T , R0 is the resistance at reference


and c is the temperature coefficient of resistance depending on the

material (Fig. 7.1.5)

There are certain classes of semiconducting materials (such as metal oxides of

cobalt, manganese and nickels) having negative coefficient of resistances. These
devices are called as thermistors for which the resistance temperature relation is nonlinear as given below;

exp t







t is a constant for a given thermistor. The practical operating ranges for the

thermistor lie approximately between 100C to 275C.

Fig. 7.1.5: Representation of a resistance temperature gauge.

Thermocouples: When two dissimilar metals are joined together to form a junction,
then an electromotive force exists between the junctions and the connecting wires that
forms the circuit. It leads to a current flow in the circuit due to the potential difference
of voltage that comes from two different sources: from contact of two dissimilar
metals at the junction temperature and from the temperature gradients along the
conductors in the circuit. The first one is known as Peltier effect which is due to the
temperature difference between the junctions. The voltage difference due to
temperature gradients along the conductors in the circuit is known as Thomson effect.
In most of the cases, the Thomson emf is quite small relative to Peltier emf and with
proper selection of conductor materials, the Thomson emf can be neglected. With this
principle, a thermocouple is prepared between two conductors of different materials
resulting two junctions p and q as shown in Fig. 7.1.6. If the junction temperatures

T1 and T2

are equal, then the emfs will balance and no current will flow. When

there is temperature difference between the junctions, then there will be net emf and a
current will flow in the circuit. When the thermocouple is used to measure a unknown
temperature, then temperature of one of the junction is known and termed as
reference junction and by measuring the net emf in the circuit, the unknown
temperature can be measured. The thermocouple circuits also follow certain laws in
addition to Seebeck effect;

Fig. 7.1.6: Schematic representation of a thermocouple junction.

Fig, 7.1.7: Illustration of thermocouple law of intermediate metal.

Law of intermediate metals: Insertion of an intermediate metal into a thermocouple

circuit does not affect the net emf, provided the two junctions introduced by a third
metal are at identical temperatures. (Fig. 7.1.7)
Law of intermediate temperatures: If a simple thermocouple circuit develops an emf
e1 when its junctions are at

T1 and

and emf of

when the junctions

are at temperatures T2 and T3 , then the same thermocouple will develop an emf

e1 e2

when the junctions are at temperatures T1 and T3 , respectively.

Depending on the combination of materials used for the conductors,

thermocouples are classified in different types. The selection of the material is based
on cost, availability, convenience, melting point, chemical properties, stability, and
output. They are usually selected based on the temperature range and sensitivity
needed. Some of the classifications are given in Table 7.1.1.
Table 7.1.1: Classifications of thermocouples




Temperature Range



41 V/C

200C to 1350C



68 V/C

-250C to 900C



55 V/C

40C to 750C



39 V/C

up to 1200C


10 V/C

up to 1800C


10 V/C

up to 1600C


10 V/C

up to 1600C


Platinum /
6% Rhodium
Platinum /
13% Rhodium
Platinum /
10% Rhodium

43 V/C

200C to 350C

Tungsten5% rhenium
Nickel18% Molybdenum

Tungsten26% rhenium
Nickel6% Cobalt

30 V/C

0C to 2320C

30 V/C

Up to 1400C

Module 7 : Lecture 2
(Incompressible Flow Part II)
Measurement of Flow Rate and Velocity
A major requirement in application areas of fluid mechanics is the determination of
flow rates and with respect to incompressible flows, they are called as flow metering.
Based on the operating principle, pressure drop, capacity, versatility, accuracy, cost,
size and level of sophistication the range varies widely For example, a crude way of
measuring flow rate of water through a household tap is through the collection of
water in a bucket and noting the corresponding time. On the other hand, a
sophisticated instrument may involve flow rate measurement through the propagation
of sound in a flowing fluid or electromotive forces when the fluid passes through a
magnetic fluid. Some of the commonly used devices are as follows;

Pitot tube and Pitot-Static probe

Obstruction Flow meters

Variable-area flow meter

Thermal Anemometers

Miscellaneous flow devices

Scattering devices

Pitot tube and Pitot-static probe

In an incompressible flow, the flow rate is generally proportional to the velocity and it
is obtained from the measured pressures of the flowing medium. The total pressure of
a flowing stream is expressed as,
p p p p V

2 pt ps


where, V is the average flow velocity and is the fluid density. If measurement is

made in such a way that the velocity of the flow is not disturbed, then the measured
pressure indicates the static pressure ps . On the other hand, if the measurement
is made such that the flow velocity of the stream is brought to rest isentropically,

the pressure obtained becomes the stagnation/total pressure pt . The difference

between these two pressures is the dynamic pressure

pd which

is the

fundamental equation for velocity measurement. The point measurements of these

two pressures are accomplished by the use of tubes (called as probes) joining the
desired location in the flow (Fig. 7.2.1). Pitot probes and Pitot-static probes are the
standard devices that
are used widely for obtaining

and pt .

Fig. 7.2.1: Schematic representation of static and stagnation pressure measurements.

A Pitot probe is a simple tube with a pressure tap at the stagnation point where
the flow comes to rest and thus the pressure measured at this point is the stagnation
pressure (Fig. 7.2.2-a). The Pitot-static probe consists of a slender double-tube
aligned with the flow and connected to a differential pressure measuring device such
as manometer (Fig. 7.2.2-b). The inner tube is fully open to the flow at the nose and
thus measures the stagnation pressure (point 1) while the outer tube is sealed at the
nose, but has the holes on the circumference of the outer wall for measuring the static
pressure (point 2). Neglecting frictional effects in Fig. 7.2.2(c), Bernoullis equation
can be applied for the point 1 and 2 to obtain the average flow velocity as given by
Eq. (7.2.1). This equation is also known as Pitot formula. The volume flow rate can
be obtained by multiplying the cross-sectional area to this velocity.

The Pitot-static probe is a simple, inexpensive and highly reliable device because
it has no moving parts. Moreover, this device can be used for velocity/flow rate
measurements for liquids as well as gases. Referring to Eq. (7.2.1), the dynamic
pressure (i.e. difference between stagnation and static pressure) is proportional to the
density of the fluid and square of the flow velocity. When this device is used for
gases, it is expected that velocity is relatively high to create a noticeable dynamic
pressure because gases have low densities.

Fig. 7.2.2: (a) Pitot probe; (b) Pitot-static probe; (c) Measuring flow velocities with a Pitot-static probe.

Obstruction Flow meters

The flow rate through a pipe can be determined by constricting the flow and
measuring the decrease in pressure due to increase in velocity at the constriction. In
order to illustrate this fact, consider an incompressible steady flow of fluid in a
horizontal pipe area

and diameter d


and diameter

D which is constricted to a flow area A2

at certain location (Fig. 7.2.3). The mass balance and Bernoulli

equation can be applied between a location before the constriction (point 1) and at
the constriction site (point 2).
balance : A V A


Bernoulli equation : p


g 2g

d 2




g 2g


Combining both the equations and solving for V2 one can obtain the expression for
velocity and flow rate

V at the constriction location.


where, p1 ,

and p2 ,

respectively and

2 p1 p2
; 14

V A2V2


are the pressure and velocities at points 1 and 2,


is the diameter ratio. It is noticed from Eq. (7.2.3) that the

flow rate through a pipe can be determined by constricting the flow and measuring the
decrease in pressure due to increase in velocity at the constriction site. The devices
based on this principle is known as obstruction flow meters and widely used to
measure flow rates for gases and liquids. Depending on nature of constriction, the
obstruction flow meters are classified as orifice, nozzles and venturimeter as shown in
Fig. 7.2.4.

Fig. 7.2.3: Flow through a constriction in a pipe.

Fig. 7.2.4: Classification of obstruction flow meters: (a) orifice; (b) nozzle; (c) venturimeter.

The orifice meter (Fig. 7.2.4-a) has the simplest design and it occupies
minimal space. It consists of a plate with a hole in the middle and this hole may be
sharp-edged/beveled/rounded. The sudden change in the flow area causes a venacontracta (minimum area) and thus leading to significant head loss or swirl. In the
flow through nozzles (Fig. 7.2.4-b), the plate is replaced by a nozzle and the flow
becomes streamlined. As a result, the vena-contracta is practically eliminated that
leads to very small head losses. The most accurate measurement device in the flow
meter category is the venturi-meter (Fig. 7.2.4-c). Its gradual contraction and
expansion prevents flow separation and swirling which minimizes the head losses.
However, it suffers irreversible losses due to the friction at the wall which is only
about 10%.
The velocity expression in the Eq. (7.2.3) is obtained by assuming no loss and
thus it is the maximum velocity that occurs at the constriction site. In reality, the
velocity will be less than this value because of inevitable frictional losses. Also, the
fluid stream will continue to contract past the obstruction and the vena-contracta is
less than the flow area of the obstruction. By incorporating these two factors, a
correction factor is introduced in the obstruction flow meters, which is measured
experimentally. The volume flow rate is then expressed by a parameter called as
discharge coefficient Cd .
V A0CdV2
The value of

2 p1 p2


A0 d


parameter and flow Reynolds number

depends on the geometrical

Re V1 D . In the range of 0.25 0.75 and 104 Re 107 , the


C of

can be approximated by the following relations;


C 0.5959 0.0312 0.184




Nozzles : Cd 0.9975




For high Reynolds number flows Re 30000, the value of


can be taken as 0.61

for orifice and 0.96 for nozzles. In the case of venturimeter, the of Cd ranges from
1.95 to 0.99 irrespective of flow Reynolds number and area ratio because this device
in intended for streamlined design.

Relative merits of venturi meter, nozzle and orifice

High accuracy, good pressure recovery and resistance to abrasion are the
primary advantages of the venturi. The space requirement and cost of the
venturi meter is comparatively higher than that of orifice and flow nozzle.

The orifice is inexpensive and may often be installed between existing pipe
flanges. However, its pressure recovery is poor and it is especially susceptible
to inaccuracies resulting from wear and abrasion. It may also be damaged by
pressure transients because of its lower physical strength.

The nozzle possesses the advantages of the venturi, except that it has lower
pressure recovery and it has the added advantage of shorter physical strength.
It is inexpensive compared with the venturimeter but relatively difficult to
install properly.

Module 7 : Lecture 3
(Incompressible Flow Part III)
Variable-Area Flow Meter
In the obstruction flow meters, the flow is allowed to pass through a reduced crosssectional area


and the corresponding pressure difference

p1 p2

is measured

by using any differential pressure measuring device. The expression for volume flow

V is given by,
V A0Cd

2 p1 p2




and larger diameters for the flow, respectively, Cd is

where, d and D are the smaller
the discharge coefficient and is the density of the fluid. It may be noted from Eq.
(7.3.1) that the pressure drop varies as square of the flow rate. In other words, if these
devices are to be used for wide range of flow rate measurements, then the pressure
measuring equipment should have capability of handling lager pressure range. By
incorporating larger pressure ranges, the accuracy of the device will be poor for low
flow rates i.e. the small pressure readings in that range will be limited by the pressure
transducer resolution. This is the major drawback of the obstruction flow devices.
One of the solutions is to use two pressure measuring systems, one for low
flow rates and the other for high flow rates. A simple, reliable and inexpensive device
used for measuring flow rates for wide ranges of liquids and gases. This device is easy
to install with no electrical connections and gives a direct reading of flow rate. It is
known as variable area flow meter and also called as rotameter/floatmeter. It consists
of a vertical tapered conical transparent tube made of glass/plastic with a float/bob
inside the tube as shown in Fig. 7.3.1. The bob is free to move inside the tube and is
heavier than the fluid it displaces. At any point of time, the float experiences three
fundamental forces; drag, buoyancy and its own weight. With increase in flow
velocity, the drag force increases and the flow velocity reduces with increase in crosssectional area in the tapered tube. At certain velocity, the float settles at a location

where enough drag

Fd is

generated to balance the weight of the bob Wb and

buoyancy force

Fb . In other words, the net

force acting on the bob is zero and

thus it is in equilibrium for a given flow rate. The degree of tapering of the tube
can be made such that the vertical rise changes linearly with the flow rate and a
suitable scale outside the tube is fixed so that the flow rate can be determined by
matching the position of float.

Fig. 7.3.1: Schematic diagram of a rotameter.

At equilibrium state, the force balance on bob can be written by the following
Fd Fb Wb


By definition, all these forces terms can be expressed in the following form;

u 2





V g;



b b

is the total volume of the bob, Ab is the frontal area of the bob, um is the

mean flow velocity in the annular space between the bob and tube, CD is the drag
coefficient, g is the acceleration due to gravity, f and

are the fluid density and

float density, respectively. Both Eqs (7.3.2 &7.3.3) can be combined to obtain the
expression for

and subsequently volume flow rate

1 2gV

um C A b b 1
b f


V Aum


V .


D ay


Here, A is the annular area, D f is the diameter of the tube at inlet, d is the
maximum bob diameter, y is the vertical distance from the entrance and a is the
constant indicating the tube taper. Since the drag coefficient depends on the Reynolds
number and fluid viscosity, special bob may be used to have constant drag coefficient.
It is also possible to decide appropriate geometrical dimensions so that a linear
relation is obtained for the expression given by the Eq. (7.3.4).
mC1 y

C1 is a constant for


Since the response of rotameter is linear, its resolution is same for both higher and
lower flow rates. The accuracy for these types of devices is typically 5%. However,
rotameters have certain drawbacks such as vertical installation and inability for
measurements of opaque fluids because the float may not be visible.
Thermal Anemometers
The thermal anemometers are often used in research applications to study rapidly
varying flow conditions. When, a heated object is placed in a flowing fluid, it tends to
lose heat to the fluid. The rate, at which the heat is lost, is proportional to the flow
velocity. If the object is heated to a known power and placed in the flowing fluid, then
heat will be lost to the fluid. Eventually, the object will reach to a temperature which
is decided by the rate of cooling. However, if the temperature of the object is to be
maintained constant, then the input power needs to be changed which is proportional
to the fluid velocity. So, the heating power becomes a measure of velocity. The
concept of using thermal effects to measure the flow velocity was introduced in late

1950s. The thermal anemometers have extremely small sensors and are useful to

measure instantaneous velocity at any point in the flow without disturbing the flow
appreciably (Fig. 7.3.2).

Fig. 7.3.2: Operating principle of a thermal anemometer.

Fig. 7.3.3: A hot-wire/hot film thermal anemometer with its support system:
(a) hotwire; (b) hot-film.

The schematic diagram of a hot-wire/hot-film probe is shown in Fig. 7.3.3. A

thermal anemometer is called a hot-wire anemometer when is the sensing element is a
wire. It is called as a hot-film anemometer if the sensor is a thin metallic film. For a
hot-wire anemometer, the sensing element has a diameter of few micron and length of
2mm. In the case of hot-film anemometer, the sensing element is of 0.1 m thick and
mounted on a ceramic support. The sensing element is usually made out of platinum,
tungsten or platinum-iridium alloy.

Fig. 7.3.4: (a) Schematic representation of anemometer measurement; (b) Anemometer feedback controlled circuit.

Both hot-wire and hot-film probes are operated using a feed-back controlled
bridge (Fig. 7.3.4) that controls the input power to the probe to maintain constant
temperature when there is a change in fluid velocity. The higher is the flow velocity,
the more will be heat transfer from the sensor and more voltage/power will be
required for the sensor. When the sensor is maintained at constant temperature, the
thermal energy remains constant. So, the electrical heating
equal to the rate of heat loss through convection

qC and

qE of

the sensor is

is often governed by

Kings law.
qC a bV

q i R



1T T

where, i is the electric current in the circuit, V is the flow velocity,

Tw T and

, T

coefficient of resistance,

is the

a and b are the calibration constants,

are the wire temperature, free stream fluid temperature and reference

temperature, respectively. With appropriate calibration, Eq. (7.3.6) can be expressed

through a close correlation between flow velocity V and voltage


E .

A, B and n are calibration constants.


One of the important applications includes the turbulence measurements

where the velocity fluctuations are important. Two or more wires at one point in the
flow can make simultaneous measurements of the fluctuating components. The
thermal anemometers have distinct advantages of measuring very high velocities
(~1000m/s) with excellent spatial and temperature resolution for liquids as well as
gases. The hot film probes are extremely sensitive to fluctuations in the fluid velocity
and have been used for measurements involving frequencies as high as 50 kHz. The
time constants of the order of 1ms can be obtained with hot-wire probes operating in

Moreover, simultaneous

measurement of



dimensional) can be done by aligning three sensors on a single probe.


Module 7 : Lecture 4
(Incompressible Flow Part IV)
Miscellaneous Measuring Devices
The preceding section covers the common types of flow meters such as obstruction
devices and rotameter. There are few additional devices that are used for specific
applications and these devices can be made such that outputs that vary linearly with
flow rate. Some of the special classes of these devices are briefly discussed here.
Electromagnetic flow meters
When a conductor is moved in a magnetic field, an electromotive force is developed
due to magnetic induction. The voltage induced across the conductor EMF while
moving right angles to the magnetic field is proportional to the velocity of the
conductor. This principle is known as Faradays law and stated by the following


where, B is the magnetic flux density (gauss), L is the length of the conductor (cm)
and V is the velocity of the conductor (cm/s). If the conductor is replaced by a
conducting fluid, then V may be replaced by flow velocity.

Fig. 7.4.1: Operating principle of an electromagnetic flow meter.

A full flow electromagnetic flow meter is a non-intrusive device consisting of a

magnetic coil that encircles the pipe containing a flowing fluid (conductive), as shown
in Fig. 7.4.1. Two electrodes are drilled and flush-mounted into the inner surface of
the pipe but do not interfere with the flow. These electrodes are then connected to the
voltmeter that measures the electric potential difference due to the flow velocity of the
conducting fluid.
Electromagnetic flow meters are best-suited for measuring flow velocities of
liquid metal such as mercury, sodium, and potassium and find applications in nuclear
reactors. They can also be used for liquids of poor conductors if they contain adequate
amount of charged particles. Flow rate measurement of corrosive liquids, slurries and
fertilizers are also possible by electromagnetic flow meters. Commercial magnetic
flow meters have rated accuracies of 0.5% to 1%.
Turbine meters
It is a rotating-wheel type magnetic flow meter which is used to measure water flows
in rivers and streams. As the fluid moves through the meter, it causes a rotation of the
small turbine wheel. A permanent magnet is encased in the rotor body such that the
change in permeability is noticed when the rotor blade passes through the pole of the
coil. The change in the permeability of the magnetic circuit produces a voltage pulse
at the output measuring terminal. The rotor motion is proportional to the volume flow
rate and is captured by an inductive coil. A flow coefficient

may be defined based on the flow rate

for the turbine meter

V and kinematic viscosity of the fluid.

V ;
f is the pulse frequency.


This particular device indicates the flow accurately within 0.5% over wide range of
flow rates.

Vortex flow meters

When a flow stream encounters an obstruction in its path, the fluid separates and
swirls around the obstruction. This leads to formation of vortex and it is felt for some
distance downstream. It is a very familiar situation for turbulent flows and a short
cylinder placed in the flow sheds the vortices along the axis. If the vortices are
periodic in nature, then the shedding frequency is proportional to the average flow
velocity. In other words, the flow rate can be determined by generating vortices in the
flow by placing an obstruction along the flow and measuring the shedding frequency.
For an incompressible fluid, a dimensionless parameter known as Strouhal number


is defined as function of vortex shedding frequency f s , characteristic

dimension of the obstruction

l and

the velocity of flow

impinging the




Fig. 7.4.2: Operation of a vortex flow meter.

A vortex flow meter works on the above principle is shown in Fig. 7.4.2. It
consists of a bluff body placed in the flow that serves as vortex generator and a
detector placed at certain distance downstream on the inner surface of casing records
the shedding frequency. A piezoelectric sensor mounted inside the vortex shedder

detects the vortices and subsequently amplified to indicate either instantaneous flow

rate or total flow over selected time interval. With prior knowledge of calibration
constant St and characteristic length dimension of the bluff body, the average flow
velocity can be obtained. The vortex flow meters operate reliably between the

Reynolds numbers of 10 to 10 with an accuracy of 1%. It is generally not suitable

for use of high viscous liquids.
Ultrasonic flow meters
When a disturbance is created in the flowing fluid, it generates sound waves that
propagates everywhere in the flow field. These waves travel faster in the flow
direction (downstream) compared to the waves in the upstream direction. As a result,
the waves spread out downstream while they are tightly packed upstream. The
difference between the number of waves in upstream and downstream is proportional
to the flow velocity. The ultrasonic flow meters operate on this principle using sound
waves in the ultrasonic range (~1MHz). Its operation mainly depends on the
ultrasound waves being reflected and discontinuities in the density. Also, solids,
bubbles and any discontinuity in the liquid will reflect the signal back to the receiving
element. So, the device requires that the liquid contains at least 25ppm (parts per
million) of particles or bubbles having diameters of 30m or more. There are few
distinct advantages of ultrasonic flow meters such as easy installation, non-intrusive
type measurement and negligible pressure drop since it does not interfere the flow.
Two basic kinds of ultrasonic flow meters include transit time and frequency shift
flow meters.
The transit time flow meter (Fig. 7.4.3-a) involves two transducers located at
certain distance l that alternatively transmits and receive ultrasonic sound waves, in

the direction of the flow as well as in the opposite direction. The travel time for each
direction can be measured accurately and the difference t can be estimated. The
average flow velocity V can be determined from the following relation;
V K l t

K is a constant


The frequency shift flow meter (Fig. 7.4.3-b) is normally known as Dopplereffect ultrasonic flow meter that measures average velocity along the sonic path. The
piezo-electric transducers placed outside the surface of the flow transmits sound
waves through the flowing fluid that reflects from the inner wall of the surface. By
capturing the reflected signals, the change in frequency is measured which is
proportional to the flow velocity.

Fig. 7.4.3: Basic principle of an ultrasonic flow meter: (a) Transit time flow meter; (b) Frequency shift flow

Laminar flow meter

It is constructed through the collection of small tubes diameter d and length l of
sufficiently small sizes (Fig. 7.4.4) so that laminar flow is ensured and the
entrance/exit losses occur within the tube assembly. Thus, the flow rate for a given
fluid viscosity

becomes direct function of pressure difference

p2 p1 .

d4pp d 4p



128 L

Since the flow is laminar, the Reynolds number

Red is

within 2000. One may

rewrite the expression of Reynolds number as follows;



d 4V

4 d 2


Combining the Eqs (7.4.5 & 7.4.6), the design selection of

l and d , for a laminar

flow meter can be set for certain range of pressure drop and flow Reynolds number.

128 Re L

p p1 p2



In contrast to obstruction flow devices, the volume flow has a linear relation with
pressure drop for a laminar flow meter. It allows the operation of this device for wide
range of flow rates for a given pressure differential, within an uncertainties of 4%.
However, being small in sizes, the laminar tube elements are subjected to clogging
when used with dirty fluids.

Fig. 7.4.4: (a) Basic principle of a laminar flow meter; (b) A laminar flow element.

Thermal mass flow meter

A direct measurement of mass flow of gases can be accomplished through thermal
energy transfer (Fig. 7.4.5). The flow takes place through a precision tube fitted with
an electric heater. Both upstream and downstream sections have externally wounded
resistance temperature detectors (RTDs) typically made out of platinum with probe
diameter of about 6mm. The first sensor measures the temperature of the gas flow at
the point of immersion while the second senor is heated to a temperature of 20C
above the first sensor. As a result, the heat transfer to the gas from the second sensor
takes place through convection which is proportional to the mass velocity

u of

the gas. The two sensors are connected to a Wheatstone bridge circuit for which
the output voltage is proportional to the mass velocity. This circuit can be
specially designed so that linearly varying output can be obtained from the circuit.
The experiment is normally performed with nitrogen and a calibration factor is
obtained for subsequent use of other gases.

It is to be noted that the mass velocity of the gas is measured at the point of
immersion. For the flow system with varying velocities, several measurements are
necessary to obtain an integrated mass flow across the channel. Velocities of the gases
in the range of 0.025m/s to 30m/s can be measured with this device within an
uncertainty level of 2%.

Fig. 7.4.5: Basic principle of a thermal mass flow meter.

Scattering Devices
All the measurement techniques discussed earlier, determine the velocity by
disturbing the flow. In some cases, the disturbance is very less (such as ultrasonic
flow meter and thermal anemometers) while in other cases (orifice, pitot probe etc.),
sufficient care to insert the measuring device to minimize the disturbances. So they
are classified as intrusive based measurements. The modern instrumentation method
used optical technique to measure the flow velocity at any desired location without
disturbing the flow. So they are called as non-intrusive based measurement and works
on the principle of scattering light and sound waves in a moving fluid. By measuring
the frequency difference between scattered and un-scattered wave, particle/flow speed
can be found. The Doppler frequency shift f is responsible for this change in the
speed and is illustrated in Fig. 7.4.6.




where, V is the particle velocity, is the wavelength of original wave before


and are the angles shown in Fig. 7.4.6. The important aspect of the

Eq. (7.4.8) is the proportionality between


and V .

f can be measured, then one

can obtain the particle speed. Since, the particle moves with the flow, so the particle
speed is equal to flow speed. Since, the laser light and ultrasonic waves have
relatively high frequencies, they are normally used for measuring flow velocity
because the Doppler shift will be only a small fraction compared to original
frequency. Laser Doppler Velocimetry (LDV) and Particle Image Velocimetry (PIV)
are the optical techniques that work on the principle of Doppler shift.

Fig. 7.4.6: Illustration of Doppler shift for a moving fluid particle.

Laser Doppler Velocimetry: It is also termed as Laser Velocimetry (LV) or Laser

Doppler Anemometry (LDA). The operating principle of LDV is based on sending a
highly coherent monochromatic light beam towards a fluid particle. The
monochromatic light beam has same wavelengths and all the waves are in phase. The
light reflected from the fluid particle wave will have different frequencies and the
change in frequency of reflected radiation due to Doppler effect is the measure of
fluid velocity. A basic configuration of a LDV setup is shown in Fig. 7.4.7. The laser
power source is normally a helium-neon/argon-ion laser with a power output of
10mW to 20W. The laser beam is first split into two parallel beams of equal intensity
by a mirror and beam-splitters. Both the beams pass through a converging lens that
focuses the beams at a point in the flow. The small fluid volume where the two beams
intersect is the measurement volume where the velocity is measured. Typically, it has

a dimension of 0.1mm diameter and 0.5mm long. Finally, the frequency information

of scattered and unscattered laser light collected through receiving lens and photodetector, is converted to voltage signal. Subsequently, flow velocity V is calculated.

Fig. 7.4.7: Schematic representation of a LDV setup.

Fig. 7.4.8: Interference of laser beams: (a) Formation of fringes; (b) Fringe lines and wavelengths.

When the waves of two beams interfere in the measurement volume, it creates
bright fringes when they are in phase and dark fringes when they are out of phase.
The bright and dark fringes form lines parallel to the mid-plane between two incident
laser beams as shown in Fig. 7.4.8(a). The spacing between fringe lines s can be
viewed as wavelength of fringes as shown in Fig. 7.4.8(b).


2V sin


It is the fundamental LDV equation that shows the flow velocity proportional to the

Particle Image Velocimetry (PIV): It is a double-pulsed laser technique used to

measure instantaneous velocity distribution in a plane of flow by determining the
displacement of particles in that plane during a short time interval. The all other
techniques such as LDV and thermal anemometry, measure the velocity at a point
while PIV provides the velocity values simultaneously throughout entire cross-section
through instantaneous flow field mapping.

Fig. 7.4.9: Experimental arrangement of a PIV system.

The PIV technique for velocity measurement is based on flow visualization

and image processing as shown in Fig. 7.4.9. The first step is to trace the flow with
suitable seed particles in order to obtain the pathlines of fluid motion. A pulse of laser
light illuminates certain region of flow field at any desired plane and the photographic
view is recorded digitally by using a video camera positioned at right angle to the
plane. After a short interval of time

t ,

the particles are illuminated again

through the laser light and the new positions are recorded. Using the information of
both the images, the particle displacement s is determined and subsequently the
of velocity

t of the particle in the plane is calculated.

A variety of laser sources such as argon, copper vapour and Nd-YAG can be used
with PIV system, depending on the requirements for pulse duration, power and time
between the pulses. Silicon carbide, titanium dioxide and polystyrene latex particles
are few categories of seed particles that are used depending on the type of fluid
(liquid/gas). In addition to velocity measurement, the PIV is capable of measuring
other flow properties such as vorticity and strain rates. The measurements can be
extended to supersonic flows, explosions, flame propagation and unsteady flows. The
accuracy, flexibility and versatility are the few distinct advantages of a PIV system.

Module 7 : Lecture 5
(Compressible Flow Part I)
The compressible flows are normally characterized as variable density flows. Pressure
gradient, variable area, heat exchange and friction are few mechanisms that can
change the density during a flow. But the conditions will vary for liquids and gases.
For instance, the pressure gradient causes predominant change in the velocity keeping
density as constant for liquids. In the other hand, the change in pressure can cause
substantial velocity and density change for gases. When the density variation is less
the 5%, the gases are still in the incompressible limit and the measurement techniques
discussed earlier can be extended to the gases as well. However, if the density
variation is substantial (more than 5%), then the measurement methods are different.
The incompressible limit fails when the Mach number M

of the flow is more than

1.3 and the flow remains subsonic till M 1 . If the Mach number of the flow is
progressively increased, then one may reach the supersonic M 1and hypersonic


limits. In other words, the variation in density is normally associated with

high speed flows.

In the compressible flow measurements category, some of the flow parameters
such as pressure, temperature are measured directly while others are calculated from
the measured parameters using gas dynamic relations. When the measurements are
performed for supersonic/hypersonic flows, a shock wave remains attached to the
body geometry across which the static pressure, temperature and density variations
are very high. Hence, the measured parameters only provide the information that
prevail after the shock. Using shock wave relations, indirect calculation can be made
to infer the desired flow parameters. Moreover, many advance measurement
techniques involve flow field visualization through density variation to get back the
information of pressure and velocity. Some of the basic compressible flow
measurement techniques are discussed here.

Measurement of Temperature
In general, the static temperature along with the pressure determines the
thermodynamic state of fluid at any instant. With compressible flow field, the
temperature and velocity of the flow is normally very high. In order to get the static
temperature, the measuring device must travel at the fluid velocity without disturbing
the flow which is quite unrealistic. So, the indirect determination of static temperature
measurement is done by using thermocouples by exposing directly into the flow or
mounting them on the wall surface. At any case, the flow disturbance due to
obstruction of temperature sensing device should be minimized.

Fig. 7.5.1: Schematic diagram of temperature measurement for compressible flows: (a) thermocouple located at the wall
surface; (b) temperature probe facing the flow.

The thermocouple is a simplest device for performing stagnation temperature

measurements of high speed gaseous streams in a compressible flow (Fig. 7.5.1). The
thermocouple located at the wall surface lies inside the viscous boundary layer at a
fixed wall (Fig. 7.5.1-a). Due to viscous effects, no-slip conditions need to be satisfied
at the wall and the flow velocity is zero at the wall surface. At the same time, if the

wall surface is insulated, then the temperature measured at the wall is called as
adiabatic wall temperature Taw . Another method is to design a probe which is
inserted into the flow by minimizing the obstruction (Fig. 7.5.1-b). It consists of a
diffuser which decelerates the velocity to a low value so that the fluid reaches the
stagnation state at the thermocouple location. If sufficient care is made for suitable
design of the probe, then it represents a thermodynamic state where the gas comes to
rest isentropically (i.e. no heat exchange between the probe and surroundings). Then,
the probe indicates the stagnation temperature T0 given by the following isentropic
V 2



2c p

is the specific heat of the fluid at constant pressure, V an Tare the


velocity and static temperature of the free stream fluid, respectively.

From, aerodynamic point of view, when the gas passes over the probe, a
boundary layer is likely to be formed due to velocity and temperature gradient. The
velocity gradient gives rise to shear stress resulting in fluid friction and heat
dissipation within the boundary layer. So, the probe will feel a temperature above the
stagnation temperature. At the same time, the temperature gradient in the boundary
layer gives rise to heat loss from the probe. The net effect of these two phenomena
has an opposite trend to cancel each other. The non-dimensional parameter, Prandtl
number Pr c p , representing the ratio of shearing effects to the heat transfer

effects is taken into account in the calculation of gas temperature. Since, the Prandtl
number for the gases is less, the heat conduction from the probe surface dominates
and the probe generally feels the temperature less than the stagnation temperature

T0 .

At the same time, if the probe is properly insulated and there is no heat

exchange through conduction by stem and radiation, then the probe temperature will
be the adiabatic wall temperature Taw . This deviation in the probe reading and

R .

isentropic stagnation temperature is expressed by adiabatic recovery factor





1 1R
T0 T


Here, the stagnation to static temperature ratio


is obtained from isentropic

relation. When the Prandtl number is unity, the adiabatic wall temperature becomes
equal to stagnation temperature. The adiabatic recovery factor expressed in Eq. (1)
applies to the case when the Prandtl number is not unity. In most cases, it is always
less than 1 and is related to adiabatic recovery factor as given below;
Laminar compressible boundary layer: R Pr


Turbulent compressible boundary layer: R Pr

0.72 for


0.9 for




With respect to measurement limitations, another deviation may arise if the

probe protrudes into the flow. Here, there are possibility of heat exchange through
conduction by the stem of the probe and radiation from the probe. So the temperature
measured by the probe is

instead of T0 or Taw . To account this fact, a correction

factor K is introduced, which is defined by the following

T p T
T0 T


In a particular flow field, if the temperature probe is designed such that heat loss from
the sensor is negligible, then the value of K is equal to R as it is the measure of
transport phenomena in the boundary layer and it gets altered by the shape of the
instrument. The various temperature discussed above in a compressible flow is shown
in Fig. 7.5.2.
When the measurement is performed for supersonic flow stream, a detached
shock is formed at a certain distance in front of the probe (Fig. 7.5.3). Since the flow
across a shock is adiabatic, the stagnation temperatures remain the same before and
after the shock. So, the temperature measurement remains unaffected.

Fig. 7.5.2: Representation of temperature trends in a compressible flow.

Fig. 7.5.3: Detached shock ahead of the measuring temperature probe in a supersonic flow.

Module 7 : Lecture 6
(Compressible Flow Part II)
Measurement of Pressure
Many pressure measuring devices used for incompressible flows can be equally
applicable for compressible flows if they are feasible for measurements in gases. They
may be grouped into manometers and pressure transducers depending on the ranges of
pressure and degree of precision. U-type liquid manometer, dial-type pressure gauge
(Bourdon tube), electrical/mechanical/optical types of pressure transducers are few
popular pressure measuring devices.
With respect to compressible flow field, the measurement concept of both
static and stagnation pressure (Fig. 7.6.1) are equally important. Both the pressures
along with the temperature can be used for calculating local flow velocity, Mach
number and density of the flowing stream. When the measurement is made in such a
way that the velocity of the flow is not disturbed, then the measured pressure indicates
the static pressure. On the other hand, if the flow is brought to rest isentropically, then
the pressure obtained, becomes the stagnation pressure.

Fig. 7.6.1: Measurement concepts of static and stagnation pressures.

Static Pressure Measurement: The wall static pressure measurement is important in

situations like inner walls of duct flows, surface of an airfoil etc. Here, a small hole is
drilled normal to the surface (commonly called as pressure tapping) so that pressure
measuring device can be connected (Fig. 7.6.2-a). In order to measure the static
pressure at any interior point in the flow, a probe may be inserted without disturbing
the flow streamlines (Fig. 7.6.2-b). The static pressure of the fluid stream over the
surface is transmitted through orifice in the plane of the flow and subsequently
recorded by the pressure measuring instrument. While measuring pressures through
probe, the position of sensing holes and the support stem is very important. The
deviation of actual and measured pressure may arise due to nose effect and stem
effect. Since, both the effects have opposite nature; it may be possible to cancel these
effects by suitable design.

Fig. 7.6.2: Static pressure measurements in compressible flows: (a) wall pressure tapping; (b) static pressure probe.

Stagnation Pressure Measurement: The stagnation pressure is an indication of

entropy level in a flowing fluid and the change in entropy is associated to the
irreversibility. When the flow from a reservoir takes place isentrpoically, the static
pressure record of the fluid in the reservoir indicates the stagnation pressure of the
fluid. This situation of measuring pressure is analogous when the flowing stream is
brought to rest isentropically. However, due to many irreversibility associated to the
flow such as shock wave and frictional effects, the stagnation pressure may not be
equal to the reservoir pressure. So, this pressure is always measured locally in the
flow field. In order to measure the stagnation pressure at any local section, a
stagnation probe is placed in the stream parallel to the flow with its open end facing
the flow as shown in Fig. 7.6.3. Thus, it allows the fluid to get decelerated

isentropically to rest through the passage. The reading in the probe gives the
stagnation pressure at the location where the nose of the probe is oriented. This device
was first used by Henery Pitot for measurement of pressure and hence named as Pitot
tube. At low Reynolds number flow, the deceleration may not be isentropic and
inaccuracy in the measurements can arise.

Fig. 7.6.3: Pitot tube for stagnation pressure measurement.

Measurements of Flow Velocity

In most of the cases, the flow velocity is obtained through simultaneous measurement
of static and stagnation pressures using a Prandtl Pitot Static probe (Fig. 7.6.4). It has
opening at the nose for stagnation pressure communications while several number of
equal size holes are made around the circumference of the probe at the location
downstream of the nose. The difference pressure gives the dynamic pressure. Further,
Bernoulli equation can be applied to calculate the flow velocity.

dp V


VdV gdz

Now, replace the integral of Eq. (7.6.1) with the isentropic relation for gases;

p c c d



where, p, and V are the pressure, density and velocity, respectively, z

is the

elevation difference, is the specific heat ratio and c is a constant. Combine Eq.

(7.6.1 & 7.6.2) and simplify to obtain the Bernoulli equation for one-dimensional
frictionless isentropic flow for compressible fluid.



Apply Eq. (7.6.3) along a stream line at the location of stagnation point and any
desired location to obtain the flow velocity.

p0V 2

V 2 p0 p


The subscripts, 0 and refers to stagnation and free stream conditions, respectively.
Had the flow been incompressible, the density term in Eq. (7.6.1) becomes constant
quantity and the stagnation and static pressure difference is expressed as follows:
p0 p

2 p0

Fig. 7.6.4: Prandtl Pitot static probe for simultaneous measurement.


Measurements for Subsonic and Supersonic Flows

The flow Mach number is one of the important parameter for subsonic and supersonic
flows. All the flow parameters and their variations are the functions of local Mach
number M

The pressure measurements are one of the common practices to

determine the Mach number. In subsonic flow, the simultaneous measurement of


p and stagnation pressures p0

using a Prandtl Pitot Static tube are made in

a similar way as shown in Fig. 7.6.4. Subsequently, the isentropic relation is used to
determine the flow Mach number.


1 1



The characteristic feature of a supersonic flow is the formation of a shock

wave. So, the introduction of a Pitot probe into the flow stream, leads to a detached
bow shock (Fig. 7.6.5). Due to this shock wave at certain distance from the
measurement location, the stagnation pressure located indicated by the probe will be
much higher than the stagnation pressure of the free stream. For the stagnation stream
lines, the curved shock is normal to the free stream and the measured value represents
the stagnation pressure downstream of the normal shock p02 . While conducting
experiment, the static pressure


the free stream (upstream of the shock) is

also measured simultaneously by any of the methods, discussed in Fig. 7.6.2.

However, the static pressure measurement must be done far upstream of the shock
so that its influence on the measurement will be minimized. The Mach number
relation connecting the static and stagnation pressure measurements is expressed by
Rayleigh- Pitot formula for supersonic flows.


1 2

2 2 1
1 M 1



The Rayleigh-Pitot formula with air as free stream is presented graphically in Fig.
7.6.6. The dynamic pressure

pd obtained from static pressure and the Mach

number is then given by the following expression.


M 2


Thus, the Mach number calculation through static and stagnation measurements gives
complete information of a supersonic flow field.

Fig. 7.6.5: Detached shock ahead of the measuring pressure probe in a supersonic flow.

Fig. 7.6.6: Mach number determination from Pitot tube measurement in a supersonic flow.

Sonic Nozzle: It is an obstruction device often used to measure high flow rates for
gases. When the flow rate is sufficiently high, the pressure differential is


expected to be large. Under this condition, a sonic flow condition is achieved at the
minimum flow area and the flow is said to be choked. Such a device is known as sonic
nozzle. In this case, the flow rate takes the maximum value for a given inlet condition.
If this inlet refers to a reservoir pressure

p0 ,

temperature T0 and the flow is


A ,

to be choked at certain area

then the pressure at this location

p can

obtained from isentropic relation,



This relation is known as critical pressure ratio for a choked nozzle. The choked mass
flow rate can be obtained by the following expression,

p0 A T


R 1


By designing the geometric parameter of a sonic nozzle, it is possible to achieve the

discharge coefficient up to 0.97 corresponding to theoretical expression of flow rate
given by the Eq (7.6.10).

Module 7 : Lecture 7
(Compressible Flow Part III)
Density Variation Techniques
The density of a flow can be calculated by measuring/determining the pressure and/or
temperature. In the case of liquids, the density decreases slightly with temperature and
moderately with pressure. All the gases at high temperatures and low pressures are in
good agreement with the perfect gas law. So, for liquids, one can neglect the
temperature effect and an empirical relation may be written for pressure

p and

while perfect gas equation can be stated ideal gases as given below;

B 1



Gases: p
where, p and are the standard atmospheric value, B and m are the dimensionless
parameters. For example, water can be approximately fitted to Eq. (7.7.1) with
B 3000 and m 7 . Since the liquids are generally treated incompressible, the
density variation is neglected. But, for compressible flows, the variation in density
can be considered as an important tool to investigate the flow patterns during the
experiments. The general principle for flow visualization for incompressible flow is to
render the fluid elements visible either by observing the motion of suitable foreign
materials added to the fluid. The other way is to use optical pattern resulting from
variation in optical properties of the fluid such as refractive index. This technique is
applicable for studying the flow pattern in compressible flows. In high-speed flows,
the density changes are adequate to make these phenomena sufficient for optical

Principle of optical instruments

There are three types of optical instruments, generally used to study the flow pattern
using the concept of density variation. They are interferometer, Schlieren apparatus
and shadowgraph. All the three instruments have the following common
characteristic phenomena.

Variation in density of a gas stream produces a corresponding change in the

refractive index of the gas.

Light passing through a gaseous stream with density gradient gets deflected in
the same manner as it does through a prism.

The refractive index in the medium of the flow field and the velocity of light
through the flow field are functions of the fluid density in the flow field. Since the
refractive index for most of the gases is close to unity, the relationship between the
refractive index

n and

the density of the gas medium



through Gadsone-Dale equation.

n 1 n1 1 n2 1


or, n 1K c


Here, the subscripts 1 and 2 denote two different conditions of the medium and the

K c

is different for if the gaseous medium is changed. Eq. (7.7.2) is

applicable for most of the gases except for very dense gases. Since, the fluid density
varies with location and time, the refractive index also follows the similar variation.
Let us consider a light ray passing through a compressible flow system
enclosed in a glass walls (Fig. 7.7.1). If the region inside the wall is same as outside,
then the light ray will follow a straight path and strike at a point S1 on the screen. If
the medium inside the glass wall has different density, the refractive index will
change and the light ray will get deflected through an angle d, striking at some
other point S at a distance dz from S . Also, there will be a time difference dt
due to the deflection of light ray, thereby covering more distance. Now, there are
three measurable quantities


dz and dt


to density variation in the

medium enclosed by glass wall and the medium outside. The operating principle
of optical instruments is based on these measured quantities. Depending on the

arrangements of the basic systems and optics used for observation of density
variation, it is possible to

get the indication of variation in density, density gradient (first derivative of density)
and change in density gradient (second derivative of density) as shown in Figs
7.7.2(a-c). In a typical flow field, an interferometer is useful in getting the density
change directly by measuring dt , the Schlieren apparatus is useful in studying the
density gradient from the information of d and the shadowgraph gives in indication
of change in density gradient by measuring dz .

Fig. 7.7.1: Deflection of light ray in a flow field.

Fig. 7.7.2: Density variations in a flow field.

Schlieren Apparatus: Consider two parallel beam of light passing through a test
section at same initial condition (Fig. 7.7.3). The test section is divided into two parts
T1 and

and a lens with focal length f is placed at a distance l from the test

section. A knife edge is kept at the focal point so that it can be moved up/down, thus
creating an obstruction to the light ray. A screen is placed at some appropriate
location such that the light rays passing through the test section can illuminate the
screen bright/dark depending on the position of the knife edge. It is also possible to

divide the screen into two parts


S2 such that S1 is the image of T1 and S2 is

the image of T2 . If two regions of the test section of the fluid are same, then the
images will also be the same. When the density of the gas in T2 change by keeping T1
as the same, the light rays passing through T2 will be show dark/bright image S2 in
the screen depending on the decrease/increase in the density of the medium. In other
words, if a disturbance is introduced in the test section, the light will be refracted so
that the image is displaced by a distance dz . Thus, the illumination on the screen is
proportional to dz which becomes the measure of density gradient of the flow. For
most of the gases, the refractive index is close to unity and the deflection is small.
Using electromagnetic theory, Schlieren equation is used to obtain the deflection
angle z in the z-direction.



3 rs
2 z

dx, where

is the specific refraction.


Fig. 7.7.3: Principle of a Schlieren system.